RESUMEN
Deep proteomic profiling of complex biological and medical samples available at low nanogram and subnanogram levels is still challenging. Thorough optimization of settings, parameters, and conditions in nanoflow liquid chromatography-tandem mass spectrometry (MS)-based proteomic profiling is crucial for generating informative data using amount-limited samples. This study demonstrates that by adjusting selected instrument parameters, e.g., ion injection time, automated gain control, and minimally altering the conditions for resuspending or storing the sample in solvents of different compositions, up to 15-fold more thorough proteomic profiling can be achieved compared to conventionally used settings. More specifically, the analysis of 1 ng of the HeLa protein digest standard by Q Exactive HF-X Hybrid Quadrupole-Orbitrap and Orbitrap Fusion Lumos Tribrid mass spectrometers yielded an increase from 1758 to 5477 (3-fold) and 281 to 4276 (15-fold) peptides, respectively, demonstrating that higher protein identification results can be obtained using the optimized methods. While the instruments applied in this study do not belong to the latest generation of mass spectrometers, they are broadly used worldwide, which makes the guidelines for improving performance desirable to a wide range of proteomics practitioners.
Asunto(s)
Proteómica , Espectrometría de Masas en Tándem , Proteómica/métodos , Humanos , Espectrometría de Masas en Tándem/métodos , Células HeLa , Cromatografía Liquida/métodos , Proteoma/análisis , Péptidos/análisis , Péptidos/químicaRESUMEN
Flow cytometry gives a unique opportunity to analyze thousands of individual cells for multiple parameters in a course of minutes. The most commonly used flow cytometry application in plant biology is estimation of nuclear DNA content. This becomes an indispensable tool in different areas of plant research, including breeding, taxonomy, plant development, evolutionary biology, populational studies and others. DNA content analysis can provide an insight into natural ploidy changes that reflect evolutionary processes, such as interspecific hybridization and polyploidization. It is also widely used for processing samples with biotechnologically induced ploidy changes, for instance, plants produced by doubled haploid technology. Absolute genome size data produced by cytometric analysis serve as useful taxon-specific markers since genome size vary between different taxa. It often allows the distinguishing of species within a genus or even different subspecies. Introducing flow cytometry method in the lab is extremely appealing, but new users face a significant challenge of learning instrument management, quality sample preparation and data processing. Not only is flow cytometry a complex method, but plant samples have unique features that make plants a demanding research subject. Without proper training, researchers risk damaging the expensive instrument or publishing poor quality data, artifacts or unreproducible results. We bring together information from our experience, key papers and online resources to provide step by step protocols and give a starting point for exploring the abundant cytometry literature.
RESUMEN
Deep learning (DL) is emerging as a successful technique for automatic detection and differentiation of spontaneous seizures that may otherwise be missed or misclassified. Herein, we propose a system architecture based on top-performing DL models for binary and multigroup classifications with the non-overlapping window technique, which we tested on the TUSZ dataset. The system accurately detects seizure episodes (87.7% Sn, 91.16% Sp) and carefully distinguishes eight seizure types (95-100% Acc). An increase in EEG sampling rate from 50 to 250 Hz boosted model performance: the precision of seizure detection rose by 5%, and seizure differentiation by 7%. A low sampling rate is a reasonable solution for training reliable models with EEG data. Decreasing the number of EEG electrodes from 21 to 8 did not affect seizure detection but worsened seizure differentiation significantly: 98.24 ± 0.17 vs. 85.14 ± 3.14% recall. In detecting epileptic episodes, all electrodes provided equally informative input, but in seizure differentiation, their informative value varied. We improved model explainability with interpretable ML. Activation maximization highlighted the presence of EEG patterns specific to eight seizure types. Cortical projection of epileptic sources depicted differences between generalized and focal seizures. Interpretable ML techniques confirmed that our system recognizes biologically meaningful features as indicators of epileptic activity in EEG.